Warning: session_start() [function.session-start]: Cannot send session cookie - headers already sent by (output started at /home/soslowski/public_html/astrophilter/about.php:2) in /home/soslowski/public_html/astrophilter/include/session.php on line 1

Warning: session_start() [function.session-start]: Cannot send session cache limiter - headers already sent (output started at /home/soslowski/public_html/astrophilter/about.php:2) in /home/soslowski/public_html/astrophilter/include/session.php on line 1
astroPHilter
Not logged in
astroPHilter v20.22
Jump to arXiv id: or
date:

What's astroPHilter?

AstroPHilter is the solution to all your astro-ph woes. It saves you time and helps you stay informed of all the latest scientific developments in areas relevant to your own research.
AstroPHilter provides a number of useful features, including its primary weighted-keyword based filtering of the daily arXiv lists, tracking of popular papers, and tracking of papers for the next astro-ph coffee discussion.

AstroPHilter was developed by Ben Barsdell during his PhD at the Centre for Astrophysics and Supercomputing, Swinburne University of Technology. Since the end of July 2016 Stefan Osłowski took over the maintenance of the code

How does it work?

The filtering performed by astroPHilter is based on weighted regular expressions. You provide a custom list of weighted keywords and phrases, and it searches the paper database for the highest-scoring matches, presenting them in order of their score from most relevant to least.

It should be noted that astroPHilter searches not only the astro-ph data, but ALL of the daily arXiv lists. It thus includes papers from all of the physical sciences, including physics, maths and computer science. It is hoped that providing researchers with a practical means of searching through papers from all of the physical sciences will improve cross-pollination and help minimise the number of wheels that are reinvented.

Development

Changes since 28th of July 2016

  • Correctly handle days when more than 1000 papers were posted to arXiv (31/03/2017)
  • Automatically skip dates which fell on the weekend
  • Navigation to the next day not present if displaying today's papers
  • Allow arXiv ID in the new format YYMM.NNNNN in the "Jump to arXiv id" field

Planned features

These are some of the features that Ben planned for astroPHilter:

  • Improved keyword matching system
  • Author matching
  • Choice of arXiv databases to search
  • Cacheing of generated pages
  • Rendering of LaTeX equations in titles
  • RSS/Atom feed output
  • Daily + Weekly versions
  • Ability to add news items and other information to coffee discussion list
  • Catch-up facility
  • Tracking popular papers within different groups
  • Statistics tracking (for quality of results and personal interest)
  • Custom one-off search facility
  • Personal library of papers
  • "Keep up to date" features
  • Local (i.e., institution) or global (i.e., public web) hosting.

I'm keeping them here, but I do not commit to implementing them. If you have some favourite features you would like to see please do let me know.

Technical info

AstroPHilter is currently written (or more correctly, hacked up) in a messy entanglement of PHP and Python, with a tiny drop of Javascript. ArXiv data are accessed in xml form using the arXiv OAI interface.

Tell me more!

Please feel free to contact Stefan Osłowski if you want to know more about astroPHilter or if you have any suggestions.